Using OpenCensus and Stackdriver Trace with Google Cloud Functions

Overview

This sample demonstrates use of the opencensus-trace package and Stackdriver Trace integration with Google Cloud Functions on the Python 3.7 runtime. The following features are covered:

setting the span context using request headers set by Google Cloud Platform
automatically tracing HTTP/gRPC calls issued by Google Cloud client libraries
exporting traces to Stackdriver Trace using background thread transport

The function downloads a text file (journal.txt) from a Cloud Storage bucket, appends a timestamp to that file and then uploads it back to the bucket.

The resulting trace data is exported to Stackdriver Trace.

Set up your local development environment

Pre-requisites

git
The Python 3.7 interpreter
pip
virtualenv
curl
The gcloud command-line tool

Get the repo

Use git to grab a local copy of the repo:

$ git clone https://github.com/stewblr/google-cloud-functions_python_trace.git
Cloning into 'google-cloud-functions_python_trace'...
remote: Counting objects: 84, done.
remote: Compressing objects: 100% (59/59), done.
remote: Total 84 (delta 42), reused 63 (delta 23), pack-reused 0
Unpacking objects: 100% (84/84), done.

Navigate to your local copy:

$ cd google-cloud-functions_python_trace/
$ ls
LICENSE  README.md  bin  main.py  requirements.txt  test_main.py

Create and activate a virtual environment

Use virtualenv to create a python3 virtual environment:

$ virtualenv --python python3 env

Activate your newly-created virtual environment:

$ source env/bin/activate

Install dependencies

Use pip to install project dependencies:

$ pip install -r requirements.txt

Set up local authentication

Running this sample locally involves interacting with live Google APIs. Calls to Google APIs must be authenticated. Fortunately, the client library handles authentication automatically once you create a service account and set the GOOGLE_APPLICATION_CREDENTIALS environment variable.

Note that your deployed function will automatically use a default service account associated with Google Cloud Functions. You may delete the service account created as part of this sample once you've completed local development.

Create a Cloud Storage bucket with an empty text file

This example downloads a text file from a Cloud Storage bucket, appends a timestamp to the body of the file, then uploads that file to the same bucket.

Use gsutil (installed as part of gcloud) to create a bucket:

$ gsutil mb gs://tracing-example
Creating gs://tracing-example/...

Use touch to create a text file:

$ touch journal.txt

Upload the file to your newly-created Cloud Storage bucket:

$ gsutil cp journal.txt gs://tracing-example
Copying file://journal.txt [Content-Type=text/plain]...
/ [1 files][    0.0 B/    0.0 B]
Operation completed over 1 objects.

Run locally

Set the GCP_PROJECT and FUNCTION_NAME environment variables:

$ export GCP_PROJECT=<YOUR PROJECT ID>
$ export FUNCTION_NAME=tracing_local

Note that Cloud Functions automatically sets the GCP_PROJECT and FUNCTION_NAME environment variables. You won't need to set these environment variables during deployment.

Execute the bin/test-local script to generate a mock request and send it to your function:

$ bash bin/test-local
Background thread started.
`journal.txt` downloaded from Cloud Storage to `/tmp/journal.txt`.
Appended `2018-08-26 10:33:05.852129` to `/tmp/journal.txt`.
`/tmp/journal.txt` uploaded to Cloud Storage.
Visit `https://console.cloud.google.com/traces/traces` to see tracing data for this request.
Sending all pending spans before terminated.
Background thread exited.
Sent all pending spans.

Visit Stackdriver Trace in your Google Cloud Console to see traces associated with your test request.

Deploy your function

Execute the bin/deploy script to deploy a function named tracing using gcloud:

$ bash bin/deploy
Deploying function (may take a while - up to 2 minutes)...done.

Test your deployed function

Execute the bin/test-deployed script to send a request over HTTP to your deployed function:

$ bash bin/test-deployed
Visit `https://console.cloud.google.com/traces/traces` to see tracing data for this request.

stewartreichling/google-cloud-functions_python_trace